No reason it shouldn't work, though I don't think anyone has touched it in
years.
If you're writing a book or something, unless you need something that
supports every obscure variant of footnote ever used in an academic paper,
you might want to look into variants of Markdown and similar less
On Mon, May 14, 2018 at 2:15 AM, Christian Lenz
wrote:
> Of Course, it will increase a bit of the complexity but it is as it is.
> This shouldn’t really not a reason why not doing this.
>
IMO, maybe it should. From an oldie but a goodie:
*Pull up the Tools | Options
is nicely connected and with all the 3d effects.
> > You give is a top level input for the master node, and it follows all
> > the dependencies?
> >
> >
> >
> > Thanks for sharing.
> >
> > Efrem McCrimon
> >
> >
> >
> >
> >
s so I wonder what would
> happen if
> you rendered it as a DAG?
>
>
>
>
> On 05/13/18 20:56, Tim Boudreau wrote:
>
>> If you took out a few apis used by almost everything - filesystems,
>> utilities, progress, data systems in particular, it's not actually
>
> There was a gap between the moment the last code snapshot for the donation
> was taken and the moment where the donation happened, so there were some
> changes in the old hg repositories during this gap.
> javascript2.model is AFAIK part of the upcoming donation 2:
>
;
> 2018-04-25 13:31 GMT+02:00 Tim Boudreau <niftin...@gmail.com>:
>
> > To follow up on this a bit, since I spent a bit of time attempting to
> > optimize this - the two big performance wins are:
> >
> > 1. Cache a byte[] and reuse it for every JAR entry (pass
I was playing with the code that used to live in contrib/callgraph -
something I wrote when I was working on graph analytics stuff for Oracle
Labs, and added d3-friendly output to it. Then ran across a library that
does 3d visualization on top of that.
Here is a graph of package dependencies for
Recollections from being on the selection committee, albeit in the Sun days
- things that work against a talk being selected:
- How to do [obvious thing one web search would tell you how to do]
- My favorite open source projects you can use
Things may be different as "code one" or with Oracle
t
>
>
> 2018-04-25 13:31 GMT+02:00 Tim Boudreau <niftin...@gmail.com>:
>
> > To follow up on this a bit, since I spent a bit of time attempting to
> > optimize this - the two big performance wins are:
> >
> > 1. Cache a byte[] and reuse it for every JAR
Hi, William,
> > but I don't know that it's really relevant to building
> > NetBeans. It is nice, and does the kind of parallelization you're
> > after, but it's more like a meta-build system - a build system for
> > running other build systems. FWIW, NetBSd/SmartOS's pkgsrc is also
> >
I could tell you whatever you want to know about Gentoo's build system, but
I don't know that it's really relevant to building NetBeans. It is nice,
and does the kind of parallelization you're after, but it's more like a
meta-build system - a build system for running other build systems. FWIW,
> The way I see it nbbuild/licenses are 'families'. And the 'License:'
> header in external/*-license.txt points to a family. A changing author line
> doesn't change the license itself / the license family.
> You and I are not the target audience for it - lawyers are. If that's what
gets
> be just compresses source code. Only a small part will have to be binary.
>
> --emi
>
> ‐‐‐ Original Message ‐‐‐
>
> On 28 May 2018 6:10 PM, Jaroslav Tulach wrote:
>
> > 2018-05-25 16:58 GMT+02:00 Neil C Smith neilcsm...@apache.org:
> >
> > > O
A "cluster" is a group of modules; it is common for there to be
interdependencies. What is in a cluster can be ad-hoc, but is usually used
to organize support for specific technologies.
http://wiki.netbeans.org/DevFaqWhatIsACluster
-Tim
On Mon, Jun 4, 2018 at 12:54 AM, Peter Cheung wrote:
>
Suites were really invented for Ant projects, and just happen to output a
cluster because a suite is usually a group of interdependent modules, so
most of the time mapping that to a cluster makes sense.
http://wiki.netbeans.org/DevFaqSuitesVsClusters
For Maven projects, if one is an
The IDE generates, on first run, a large random number which identifies
that install. No personal data is involved. That number is included in
HTTP requests to fetch updates.
-Tim
On Mon, Jun 4, 2018 at 11:03 AM, Daniel Gruno wrote:
> On 06/04/2018 09:56 AM, Jiří Kovalský wrote:
>
>> I do
nto a suite?
> >
> > Regards
> > Peter
> >
> >
> >
> > Am 05.06.2018 um 07:59 schrieb Tim Boudreau:
> >
> >> Suites were really invented for Ant projects, and just happen to output
> a
> >> cluster because a suit
;>>
>>> Kind regards
>>> Peter
>>>
>>>
>>>
>>> Am 05.06.2018 um 08:17 schrieb Geertjan Wielenga:
>>>
>>>> Tutorial: https://platform.netbeans.org/tutorials/nbm-maven-quickstart
>>>> .html
>>>>
FWIW, I have run against the same ~/.netbeans/dev since 2015. I think
there have been a few releases in that time. That said, there are a lot of
NetBeans modules I do not use.
The directory per release thing has always seemed a bit draconian.
It might be preferable to do this instead:
- On
The usual answer is "have a test that fails". What is a test that will
fail if something was forgotten?
--
http://timboudreau.com
I don't know that I can be much help, but the CVS client started as a
standalone project when the sources were in CVS hosted by CollabNet - it
was a github-like setup where everything was under separate source control.
I don't know what year it was that we switched to self hosting with
Mercurial,
> "One or more lists are temporarily unavailable"
>
> On 14.02.2018 5:03, Tim Boudreau wrote:
> > I don't know that I can be much help, but the CVS client started as a
> > standalone project when the sources were in CVS hosted by CollabNet - it
> > was a github-like
On Wed, Feb 14, 2018 at 1:15 PM, Martin Entlicher <
martin.entlic...@oracle.com> wrote:
> The sources are still in the NetBeans HG repo at
> http://hg.netbeans.org/core-main/file/tip/lib.cvsclient
>
> It was not developed for a long time already as Tim said and the CVS
> support was removed from
I recall there was some plumbing - written I think by Petr Hrebejk and
Honza Chalupa - that handles stack-trace-matching, so that if the same
stack trace is seen a threshold number of times from different users
reporting them (and with special cases for things like
StackOverflowException and
It's been a long time since I worked on the code involved, so my
recollection may be a little fuzzy, but this may help point you in a
direction or two:
There is a general problem that NetBeans supports, or at least used to
support, both single window pseudo-SDI mode and a multiple windows mode.
When they were written (I believe by Trung Ducks Tran in '99 or so), the
reason was to avoid opening a console window on Windows.
No idea if that's still an issue.
-Tim
On Sat, Feb 24, 2018 at 9:20 PM Scott Palmer wrote:
> I wonder why any native launchers aren’t simply
I'm frequently annoyed by this too. Has a bug been filed?
-Tim
--
http://timboudreau.com
Find the file (via data object most likely) in the current selection, then
ask project manager what project owns it.
Note multiple files may be selected in different projects or a file that
belongs to no project may be selected.
-Tim
On Mon, Jul 30, 2018 at 3:06 PM georgetrou...@gmail.com <
>
> Resurrecting this thread, since Tim Boudreau recently Mavenized the
> contrib repo:
>
> https://github.com/timboudreau/netbeans-contrib
I sent a note to this list about that on Monday, but got no response at
all. Wondering if it went into a black hole,
On Wed, Jul 25, 2018 at 8:34 AM Geertjan Wielenga
wrote:
> PS: Just for the record, Oracle doesn't relicense anything at all to
> Apache. Instead, Oracle donates code to Apache, and we together, here in
> Apache NetBeans community, do the relicensing.
Great. So what will it take to make that
l I +1 that friend API? I might be more
> > > inclined not to.
> > > Which means in practice we just stop using friends APIs. But... the
> need
> > > friend APIs cover is still there. What will we use then? It might be
> > > something even worse (like reflection,
On Wed, Jul 25, 2018 at 1:11 PM Emilian Bold
wrote:
> I think C++ is much more important.
>
+1. *That* contrib will get handled is more important than that it happen
immediately (nice as that would be).
-Tim
I don't know that this is the answer, but I know there used to be a
mechanism for marking tests that fail randomly or were known to fail but
hadn't been fixed, so the test suite didn't count them, or flagged an
unexpected pass. Maybe that's being used?
-Tim
On Wed, Jul 25, 2018 at 12:54 PM
I'll pitch a couple of things in here:
1. Part of the problem with friend apis is your that the term gets
overloaded for two different concerns: A. APIs undergoing stabilization,
where the expectation is that a stable api will emerge, and B. Dependencies
which are never intended to produce an
Hey there - I wasn't intending to blast anybody, just act as a little
corner of institutional memory, since there are a lot of people who weren't
around then and when issues are revisited, it helps to know what was
learned the last time around. Sorry if the "uh oh" that went off on my head
led me
If anyone shows up on the mailing list complaining about performance, we
> can tell them to get rid of the image and copy the bundle over.
>
99.9% of people who have that experience will NOT show up onn the mailing
list here. They'll simply use something else - perhaps blogging about the
a bad rep for desktop software isn't an excuse - it's not
like we have to make things crappier to fit in.
-Tim
On Tue, Aug 14, 2018 at 4:23 PM Will Hartung wrote:
> On Tue, Aug 14, 2018 at 1:04 PM, Tim Boudreau wrote:
>
> > If anyone shows up on the mailing list complaining about per
Perhaps an automated warning a couple of days before closure?
-Tim
On Fri, Aug 17, 2018 at 2:35 AM Jaroslav Tulach
wrote:
> Dne Pá 17. srpna 2018 06:30:28, Peter Steele napsal(a):
> > So my thoughts on this would be to decide on a more formal PR approach
> and
> > apply it to this situation
>
>
> Regards,
>
> Scott
>
> > On Aug 12, 2018, at 12:19 AM, Tim Boudreau wrote:
> >
> > This debate was had once about 14 years ago - and the decision to go with
> > .pkg installers on Mac OSX was made for this reason: There were a lot of
> > "NetBeans
>
> Clusters form the distribution and functional unit of Apache NetBeans.
> With
> the introduction of ergonomics they are becoming quite visible even for
> end
> users. In no way I'd dare to call them arbitrary.
>
Exactly. They are a unit of binary *packaging*. Which makes them a
strange
Given that clusters occasionally get split, I'm not sure cluster is the
right splitting criterion - it just cements the current layout in place.
If the problem being solved is just to make the github home page pretty,
just moving the current root contents into a src/ directory would do the
job.
Describe more how git fits into the picture please.
If you're just looking for a UI to show sets of files that were edited, and
maybe group diff, that doesn't seem like it would need git.
Or you want to somehow make local history contiguous with git history? (My
intuition says this would be a
Sounds like you would just need to modify the action and the enablement
logic, and then give the UI the ability to show history for multiple files
in one ui.
-Tim
On Wed, Aug 22, 2018 at 3:58 AM mark stephens
wrote:
>
>
> On 22 Aug 2018, at 08:29, Christian Lenz christian.l...@gmx.net>>
>
> I can be prepared on my side when I use something unstable and can adopt
> the change.
>
You can. If you're writing an IDE module, your users can't. They just
upgrade the IDE and find something doesn't work anymore, say "this sucks"
and go download Eclipse.
There are friend APIs in
>
> > So there's no reason the module system itself couldn't just discard
> > "friend" status for any friend API and allow any module to depend on it,
> > where the running release version is >= the first-appeared-in-release + 2
> > (or whatever number). So, a friend API that appears in release 9
>
> I agree with Emilians comments (have a look at the end of this mail) and
> I want to add that I
> like it to start with Friend-only API because then a potential user must
> become active to use
> it. The user asks me to be added on the friend-list and I get his e-mail
> and I can ask him for
On Sun, Jul 22, 2018 at 7:18 PM Emilian Bold
wrote:
> In the 1st idea if everything is static you don't need lookup listeners
> perhaps, you don't need module.install, etc. Probably doesn't gain you much
> except realise NetBeans would work without our specific module system.
Create a maven
>
> While I appriciate the idea of automatically propagating API state from
> instable/friend to stable after some time, there may also be less APIs
> as a result.
We have fewer APIs already - lots of friend APIs, almost no new public
APIs. It has been that way for years. Continuing to do what
Add --fontsize 24 (or whatever size works) to your netbeans.conf and turn
off zoom.
Of course fonts will look fuzzy when zoomed - you are making the logical
pixels take up more than one physical pixel.
There are a couple of system properties that may be useful to tell AWT as
well.
I have been
The original idea of friend apis was to allow development of an api with
the understanding that nobody gets things perfect the first time, so
clients v could be developed but at the risk of dealing with breakage.
The point was for the api to be in friend mode until it matures, at which
point it
Pull requests are good; automation is better. NetBeans knows its own
release version. The only (potentially) missing information is what
release the friend API first appeared in.
So there's no reason the module system itself couldn't just discard
"friend" status for any friend API and allow any
IMO, if we wanted to do this and be future-proof, the thing to do would be
to convert the icons to SVG or some similar vector format and update the
icon loading code in ImageUtilities to use it.
There are some tools - particularly potrace - that can assist in initial
conversion, but deal in
c. This shortcut
> only Woks, when I select a Project. If I’m in the Editor, it will set a
> bookmark or open the bookmark window. So there is smth like a context
> sensitive way to implement shortcuts but I don’t know the logic at all.
> >
> > My 2 cents
> >
> >
First - we're talking 8 years from now. Eight years is a long time. That
doesn't mean don't be ready for it, but it does mean don't panic.
Secondly, what it means is that Oracle doesn't plan to develop AWT and
Swing after that point. That doesn't mean nobody will or could.
That said, the way
See if the downloaded jars actually match if you download it and compare it
to your local one using shasum. If they don't match, unzip them and see
what's different.
-Tim
On Mon, Mar 12, 2018 at 8:48 PM Christian Bourque <
christian.bour...@gmail.com> wrote:
> Hi all,
>
> I'm currently trying
>
> We could try, I guess, if the website can handle the traffic (we may be
> talking about 100s of GB per month just for the catalog).
>
I know I'm jumping in late here, but a few thoughts...
1. HTTP has pretty robust caching mechanisms - is the update center making
use of them? A
Your problem is most likely your operating system's default file encoding
here (perhaps MacRoman?). The IDE is assuming that process output is
whatever your operating system's default encoding is, which is the right
assumption, since that *is* what command-line utilities will output. It
happens
r define it.
>
> Victor Williams Stafusa da Silva
>
>
> 2018-04-20 20:00 GMT-03:00 Tim Boudreau <niftin...@gmail.com>:
>
> > Your problem is most likely your operating system's default file encoding
> > here (perhaps MacRoman?). The IDE is assuming that process
Okay, legacy web application - what appserver?
Seems like the bug here is that, if NetBeans is starting the application
server, that it should detect (from configuration files or wherever) the
encoding that application server will use for logging (or know what it
is). That would be an actual fix
To follow up on this a bit, since I spent a bit of time attempting to
optimize this - the two big performance wins are:
1. Cache a byte[] and reuse it for every JAR entry (pass in a lambda to
read() rather than get out a Map)
2. Maven's DefaultScanner will pass the indexer *every
Just taking five minutes to glance at ClassDependencyIndexCreator, I see a
bunch of possible optimizations:
1. For each JAR file, it creates a Map of
className:classBytes, and then iterating it and creating a ClassFile over
each one. Garbage-collection wise, it would be
I've noticed a few bugs I filed where I included patches, *and* the patches
were accepted and the bugs were marked fixed in Bugzilla, where the code
changes are not in the Apache source code.
Digging up the exact issue numbers in Bugzilla when you have a bugzilla
history as large as mine is not
Performance wise you're definitely better off building once before opening
- scanning class files is much faster than sources.
-Tim
On Sun, Apr 1, 2018 at 12:16 PM Chuck Davis wrote:
> So I decided to look at NB code base. Followed the "bootstrap" section
> here:
FWIW, there *is* an HTTP server running there, but it drops the connection
as soon as the request line is completed:
tim@otis ~ $ telnet hg.netbeans.org 80
Trying 137.254.60.37...
Connected to hg.netbeans.org.
Escape character is '^]'.
GET / HTTP/1.1
Connection closed by foreign host.
Personally, I think revising history is somewhat evil, and disable it on my
own git servers. Usually it gets used just to turn a messy real history
into a lovely shining jewel that reflects well on the developer, but that's
mostly about ego and looking like you never make mistakes. There might
I usually set a 2-4gb heap size - I tend to naturally factor things into
small libraries and maintain a mature framework, so having 70 open maven
projects is not unusual for me.
The Achilles heel of large heaps is garbage collection. The more memory to
scan, the longer the pause. If there are
Out of curiosity, how many donations are left?
-Tim
On Tue, Oct 30, 2018 at 6:04 AM Geertjan Wielenga
wrote:
> Hi all,
>
> Oracle has handed over the grant document and ZIP providing the 3rd code
> donation to Apache. Apache is evaluating the grant document and we will
> proceed with the
I'd strongly recommend going with Antlr-4 - it is very nice.
There is one impedance mismatch to be aware of, to save a few hours of
cursing both Antlr and NetBeans' lexer infrastructure: Antlr may give you
an EOF token that is non-empty - check it, or you won't consume all the
characters from the
Maven's 'system' scope looks like it is good for something, but it rarely
(possibly never) is.
The few times I thought that was what I wanted, what I actually wound up
using was a entry where the URL file: protocol with a relative
path to some folder where the needed JAR files were pre-copied
>
> But may be that's not enough, I've raised it to 4092m. May be that'll
> perform better.
>
Turn it DOWN not up. Your Java heap is in your swap file, and has to be
swapped back in to memory for the garbage collector to scan it. So your
machine is furiously thrashing bytes off disk and into
What problem are you trying to solve?
Or, rather, what does this solve that editing netbeans,conf (by hand or
programmatically) it a module that calls System.setProperty()?
-Tim
On Thu, Sep 27, 2018 at 8:37 PM Laszlo Kishalmi
wrote:
> Hi all,
>
> What do you think on the idea to create a
You may want to survey modules on Github that use implementation
dependencies or use Uenta to bypass them. For example, I've been tweaking
the rust module from github, which does tag with a couple of csl modules.
Those should count toward "friends".
-Tim
On Fri, Sep 21, 2018 at 8:01 PM Laszlo
I've been thinking for at least a decade and a half that javahelp needs to
die. It's basically a clone of the Windows 3.1 help system. Evidence was,
last I knew, that it was rarely used by real users.
Online help would, IMHO, be fine in this era.
-Tim
On Tue, Sep 18, 2018 at 6:15 AM Peter
An extension API is something that could be bolted on top of CSL pretty
simply as another layer. For example, an implementation of the semantic
highlighting interface that delegates to instances it finds in some lookup.
And repeat that pattern for everything else.
That would be straightforward to
Generate HTML files. Publish them online. Let modules include a help URL
and when herp is invoked, find out via the classloader.
A netty based web server can run in an 8mb memory footprint. Have an
"offline help" module that bundles them and a tiny web server. For third
party modules, have the
On Tue, Sep 25, 2018 at 1:04 PM Antonio wrote:
> Hi Kenneth,
>
> I don't think there's any security related problem here. The Apache
> Mirror System, for instance, uses "http" frequently. IMHO there's no
> need to encrypt files that are publicly available for everyone to see.
Nonsense. There
I've been wondering about this for more than a year, and it happened today
while working on a NetBeans module project. It happens once every few
months - you're working on some project, and go to build it and discover
that *something* has added this to your pom.xml:
On Wed, Dec 26, 2018 at 11:35 PM venkatram.akkin...@gmail.com <
venkatram.akkin...@gmail.com> wrote:
> I have the following string which I am trying to read through
> LexerInput.read().
>
> quote
> టోకెన్
> quote
>
> ట ో క ె న ్
>
> 5 (quote) + 6(టోకెన్) + 5 (quote) + 2 new line characters that
ays try to replace the existing CharStream implementation with
> mine and see what changes.
>
> -- Eirik
>
> -Original Message-
> From: Tim Boudreau
> Sent: Friday, November 30, 2018 6:29 PM
> To: dev@netbeans.incubator.apache.org; d...@netbeans.apache.org
> Subjec
le
newbies are.
-Tim
>
> Gj
>
>
> On Sun, Nov 25, 2018 at 9:53 PM Tim Boudreau wrote:
>
> > There's a UI principle Joel Spolsky expounded years ago: Never force the
> > user into a choice they don't care about.
> >
> > A lot of users of NetBeans are ne
On Wed, Nov 28, 2018 at 1:37 AM Emilian Bold wrote:
> I was just thinking yesterday that I would love some (shell) script or
> something to add / remove plugins into an existing build folder (or
> user dir).
Here you go. Well, nodejs but invokable like a shell script:
There's a UI principle Joel Spolsky expounded years ago: Never force the
user into a choice they don't care about.
A lot of users of NetBeans are new to Java or new to programming
altogether. They don't know what Ant, Maven or Gradle are, and won't
understand what they are for a long time.
We
On Thu, Dec 27, 2018 at 5:18 PM venkatram.akkin...@gmail.com <
venkatram.akkin...@gmail.com> wrote:
> First of all, thank you for the detailed answer.
>
> > LexerInput returns a primitive int. It cannot return null.
>
> Yes sir, so I get a integer value zero. EOF as you know would be a -1. I
>
You compiled some of it with a newer JDK's javac (or source level), and
some with an older one. My guess is you started a build, some of the
tooling got compiled, then in failed with an error about the JDK being too
new, and so you switched JDKs and tried to keep going.
Hmm, OR, you are using a
getDeclaredConstructor() will find non-public constructors - that is not a
1:1 replacement.
I wouldn't litter the codebase with code to do all the needed checks if you
replaced it - rather add a " newInstance(Class, ClassLoader)" method
somewhere.
-Tim
On Fri, Mar 29, 2019 at 3:44 PM Brad
On Thu, Apr 4, 2019 at 7:40 AM Laszlo Kishalmi
wrote:
> Dear all,
>
> I would like to ask your opinion about changing the user and cache-dir
> with every release.
>
> We have been pretty much compatible with our plugins and code since 8.2
That's great, but only until somebody breaks something,
I did most of the icons in 1999 (a few of them still exist in core as tree
icons for nodes that are not typically shown anymore); in 2000 they were
taken over by Sun's Human Interface Engineering team, and everything was
converted to the (awful) "flush 3d" metal look and feel look. Circa 2004 we
So, I've got some stuff working for Antlr grammars, where you can implement
one method and one annotation, and register a formatter for an Antlr
grammar. That all works.
Getting the editor to behave right when replacing the document text is
another matter.
The formatting API is
" The best way to win that entire game is not to play." - I really wish I
> had that choice. But given that we're running a pretty big team with
> significant code being churned out, only choice I have is to ensure the
> logs are structured.
>
I wasn't saying don't do it, I was saying that every
Experience from actually trying to write a plugin that embedded a font: You
can't in any useful way for the editor. You can programmatically install a
font on Java for use by the JVM, but the resulting Font object then has to
be passed around by reference. Editor colorings look up fonts by name
I think it's a question of user experience, and this need not be an
either/or choice with the right design and supporting plumbing. How about:
Project opens fast. Dependencies are not immediately loaded - they will be
loaded lazily. Dependency nodes and such show a loading state until that
has
FWIW, I've played with *programmatic* conversion of icons to SVG. It is not
exactly practical (at least without, say, a bunch of deep learning stuff).
Basically the task is a variety of pattern recognition problems and
optimization problems - first recognizing WHAT to draw, and then how to do
that
ed
graphic designer I've worked with.
-Tim
On Sat, Apr 6, 2019 at 3:09 PM Geertjan Wielenga
wrote:
> On Sat, Apr 6, 2019 at 7:50 PM Tim Boudreau wrote:
>
> > I did most of the icons in 1999 (a few of them still exist in core as
> tree
> > icons for nodes that are not ty
If we're talking about Windows, there is an issue I went a few rounds with
engineers at Microsoft with over 15 years ago:
What really happens on Windows when you stat a file (say, with
File.isDirectory, checking time stamp, etc.), the calling thread is put to
sleep, the job is put on a queue, the
Back in 2003/4, I wrote the tab control for the NetBeans 3.6 window system,
which is still in use - the thing JTabbedPane should have been if it had
been written with the needs of applications like NetBeans in mind (i.e.
model driven, not using the AWT hierarchy as its model, capable of complex
to have a way to do this:
>
> https://www.lifewire.com/use-image-trace-in-adobe-illustrator-cc-2017-4125254
>
> Cheers,
> Antonio
>
> [1]
>
> http://mail-archives.apache.org/mod_mbox/netbeans-dev/201904.mbox/%3cCA+qecRNnE=L49v5t46q_LVc=rpTqJD3U7zt4-0DAroG=x6h...@mail.gmail.com
96 matches
Mail list logo